Search Images Maps Play YouTube News Gmail Drive More »
Sign in
Screen reader users: click this link for accessible mode. Accessible mode has the same essential features but works better with your reader.

Patents

  1. Advanced Patent Search
Publication numberUS20060179438 A1
Publication typeApplication
Application numberUS 11/276,013
Publication dateAug 10, 2006
Filing dateFeb 9, 2006
Priority dateFeb 9, 2005
Also published asUS7870552, US8341634, US8739173, US20110107343, US20140019978, WO2006086575A2, WO2006086575A3
Publication number11276013, 276013, US 2006/0179438 A1, US 2006/179438 A1, US 20060179438 A1, US 20060179438A1, US 2006179438 A1, US 2006179438A1, US-A1-20060179438, US-A1-2006179438, US2006/0179438A1, US2006/179438A1, US20060179438 A1, US20060179438A1, US2006179438 A1, US2006179438A1
InventorsDavid Jackson
Original AssigneeCluster Resources, Inc.
Export CitationBiBTeX, EndNote, RefMan
External Links: USPTO, USPTO Assignment, Espacenet
System and method of providing a fixed time offset based dedicated co-allocation of a common resource set
US 20060179438 A1
Abstract
Disclosed are a system, method and computer-readable medium relating to managing resources within a compute environment having a group of nodes or computing devices. The method comprises, for each node in the compute environment: traversing a list jobs having a fixed time relationship, wherein for each job in the list, the following steps occur: obtaining a range list of available timeframes for each job, converting each availability timeframe to a start range, shifting the resulting start range in time by a job offset, for a first job, copying the resulting start range into a node range, and for all subsequent jobs, logically AND'ing the start range with the node range. Next, the method comprises logically OR'ing the node range with a global range, generating a list of acceptable resources on which to start and the timeframe at which to start and creating reservations according to the list of acceptable resources for the resources in the group of computing devices and associated job offsets.
Images(5)
Previous page
Next page
Claims(21)
1. A method of managing resources within a compute environment comprising a group of computing devices, the method comprising:
A. for each node of a group of nodes within the compute environment:
(1) identifying at least two jobs wherein each of the at least two jobs will consume disparate types of resources within the compute environment;
(2) identifying a listing of available time frames for each of the at least two jobs which are available for reservation in order for the at least two jobs to consume their respective type of resource, the available time frames including a fixed time relationship between each of the at least two jobs;
(3) generating utilizing the listing of available time frames a start range;
(4) generating a node range based on the start range for a first job of the at least two jobs;
(5) for all subsequence jobs, logically AND'ing the start range with the node range; and
B. based on at least the logical AND'ing of the start range with the node range, generating a list of acceptable resources on which to start and a timeframe at which to start.
2. The method of claim 1, wherein the parameters associated with each job comprise at least one of duration, resource requirements, allocated resources and actions.
3. The method of claim 1, wherein the fixed time relationship between each of the at least two jobs results in at least one of job overlap, job juxtaposition or an offset-based relationship between jobs
4. The method of claim 3, wherein the fixed time relationship between each of the at least two jobs is established by specifying and enforcing the fixed time offset relationship between each of the jobs.
5. The method of claim 1, wherein generating a node range based on the start range for a first job of the at least two jobs further comprises:
shifting the start range in time by a specified job offset and for the first job, converting the shifted start range into a node range.
6. The method of claim 1, wherein processing the start range with the node range further comprises:
for all subsequent jobs, logically AND'ing the start range with the node range; and
once all jobs and nodes have been processed by logically AND'ing the start range with the node range, logically OR'ing the node range with a global range, wherein the global range reflects the timeframes and resource quantities available to an overall grouping of the at least two jobs where all jobs can simultaneously consume their requested resources at the specified offsets on the same set of allocated resources.
7. The method of claim 6, further comprising:
generating a list of acceptable resources on which to start and a timeframe at which to start; and
reserving resources according to the list of acceptable resources in the group of computing devices.
8. The method of claim 1, further comprising:
determining whether a job of the least two jobs has a highest priority, and if yes, providing first access to all currently available resources to the highest priority job.
9. The method of claim 1, further comprising creating reservations according to the list of acceptable resources for the resources in the group of computing devices.
10. A method of managing resources within a computing environment comprising a group of computing devices, the method comprising:
A. for each resource in the compute environment:
(1) traversing a list jobs having a fixed time relationship, wherein for each job in the list, the method comprises:
a. obtaining a range list of available timeframes for each job;
b. converting each availability timeframe to a start range;
c. shifting the resulting start range in time by a job offset;
d. for a first job, copying the resulting start range into a node range;
e. for all subsequent jobs, logically AND'ing the start range with the node range;
B. logically OR'ing the node range with a global range; and
C. generating a list of acceptable resources on which to start and the timeframe at which to start.
11. The method of claim 10, further comprising:
determining whether a job has a highest priority, and if yes, providing first access to all currently available resources.
12. The method of claim 10, wherein the global range reflects the timeframes and resource quantities available to the overall grouping of jobs where all jobs can simultaneously consume their requested resources at the specified offsets on the same set of allocated resources.
13. The method of claim 10, further comprising:
D. creating reservations according to the list of acceptable resources for the resources in the group of computing devices and associated job offsets.
14. The method of claim 10, wherein each resource in the compute environment referenced in step A is a node.
15. The method of claim 10, wherein the fixed time relationship between each of the at least two jobs results in at least one of job overlap, job juxtaposition or an offset-based relationship between jobs.
16. A system for managing resources within a computing environment comprising a group of computing devices, the system comprising:
A. a module configured, for each resource in the compute environment, to:
(1) traverse a list jobs having a fixed time relationship, wherein for each job in the list, the module further configured to:
a. obtain a range list of available timeframes for each job;
b. convert each availability timeframe to a start range;
c. shift the resulting start range in time by a job offset;
d. for a first job, copy the resulting start range into a node range;
e. for all subsequent jobs, logically AND the start range with the node range; and
B. a module configured to logically OR the node range with a global range; and
C. a module configured to generate a list of acceptable resources on which to start and the timeframe at which to start.
17. The system of claim 16, wherein the global range reflects the timeframes and resource quantities available to the overall grouping of jobs where all jobs can simultaneously consume their requested resources at the specified offsets on the same set of allocated resources.
18. The system of claim 16, further comprising:
D. a module configured to create reservations according to the list of acceptable resources for the resources in the group of computing devices and associated job offsets.
19. A computer-readable medium storing instructions for storing instructions for controlling a computing device to perform the steps managing resources within a computing environment comprising a group of computing devices, the instructions comprising:
A. for each resource in the compute environment:
(1) traversing a list jobs having a fixed time relationship, wherein for each job in the list, the method comprises:
a. obtaining a range list of available timeframes for each job;
b. converting each availability timeframe to a start range;
c. shifting the resulting start range in time by a job offset;
d. for a first job, copying the resulting start range into a node range;
e. for all subsequent jobs, logically AND'ing the start range with the node range;
B. logically OR'ing the node range with a global range; and
C. generating a list of acceptable resources on which to start and the timeframe at which to start.
20. The computer-readable medium of claim 19, wherein the global range reflects the timeframes and resource quantities available to the overall grouping of jobs where all jobs can simultaneously consume their requested resources at the specified offsets on the same set of allocated resources.
21. The computer-readable medium of claim 19, the instructions further comprising:
D. creating reservations according to the list of acceptable resources for the resources in the group of computing devices and associated job offsets.
Description
PRIORITY DATA

This non-provisional patent application claims priority to U.S. Provisional Application No. 60/651,269, filed Feb. 9, 2005, the contents of which are incorporated herein by reference.

BACKGROUND OF THE INVENTION

1. Field of the Invention

The present invention relates to a resource management system and more specifically to a system and method of providing a fixed time offset-based dedicated co-allocation of a common resource set in the context of managing resources within a compute environment such as a grid or a cluster.

2. Introduction

Managers of clusters desire maximum return on investment often meaning high system utilization and the ability to deliver various qualities of service to various users and groups. A cluster is typically defined as a parallel computer that is constructed of commodity components and runs as its system software commodity software. A cluster contains nodes each containing one or more processors, memory that is shared by all of the processors in the respective node and additional peripheral devices such as storage disks that are connected by a network that allows data to move between nodes. A cluster is one example of a compute environment. Other examples include a grid, which is loosely defined as a group of clusters, and a computer farm which is another organization of computer for processing.

General background information on clusters and grids may be found in several publications. See, e.g., Grid Resource Management, State of the Art and Future Trends, Jarek Nabrzyski, Jennifer M. Schopf, and Jan Weglarz, Kluwer Academic Publishers, 2004; and Beowulf Cluster Computing with Linux, edited by William Gropp, Ewing Lusk, and Thomas Sterling, Massachusetts Institute of Technology, 2003.

Often a job will include several requirements and may be dependent upon another job. An illustrative job may be that every night a television station needs to perform a weather study for its weather reports. The station may contract with an owner or operator of a cluster or a grid for computing time from 2:00 AM to 4:00 AM to utilize compute environment resources to run the analysis. The analysis may involve reserving the resources and include several jobs that require different kinds of resources within the compute environment. For example, a first job may require several hours of processing time, but also be dependent upon a second job to perform data staging or preparation of data prior to the beginning of the first job. Where these two jobs are inter-related and require the consumption of several different types of resources within a cluster or a grid, there is needed in the art for way to maximize the use of resources while being able to process jobs requiring disparate but interrelated resources.

SUMMARY OF THE INVENTION

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The features and advantages of the invention may be realized and obtained by means of the instruments and combinations particularly pointed out in the appended claims. These and other features of the present invention will become more fully apparent from the following description and appended claims, or may be learned by the practice of the invention as set forth herein.

The invention herein relates to systems, methods and computer-readable media for optimizing the resources used in a compute environment such as a cluster or a grid. The invention enables an evaluation of all resources in an instance within a single node and then viewing the resources at a cluster level or as an aggregate of resources level (several clusters). The method involves considering a single node and stepping through all jobs or all requirements and/or resource requests on that given node and determining the resource availability for each resource requester within that node. Then the system arrives at a composite availability for that node before it analyses a subsequence node.

The method embodiment of the invention relates to at a method of managing resources within a computing environment comprising a group of computing devices. The method comprises, for each node in the compute environment: (1) traversing a list jobs having a fixed time relationship, wherein for each job in the list, the following steps occur: a. obtaining a range list of available timeframes for each job; b. converting each availability timeframe to a start range; c. shifting the resulting start range in time by a job offset; d. for a first job, copying the resulting start range into a node range; and e. for all subsequent jobs, logically AND'ing the start range with the node range. Next, the method comprises logically OR'ing the node range with a global range, generating a list of acceptable resources on which to start and the timeframe at which to start and creating reservations according to the list of acceptable resources for the resources in the group of computing devices and associated job offsets. The reservation will typically entail the co-allocation of resources according to the analysis done as disclosed herein.

In addition to the method embodiment, the invention includes embodiments such as a system (whether it is a single computing device or a group of compute devices in a compute environment) and a computer-readable medium storing instructions for controlling a computing and/or a compute environment as set forth in the method.

As can be seen from the above, the invention involves, for interrelated jobs that consume disparate resources, performing an analysis of the compute environment in order to maximize the establishment of reservation of compute resources to process the jobs.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to describe the manner in which the above-recited and other advantages and features of the invention can be obtained, a more particular description of the invention briefly described above will be rendered by reference to specific embodiments thereof which are illustrated in the appended documents and drawings. Understanding that these drawings depict only typical embodiments of the invention and are not therefore to be considered to be limiting of its scope, the invention will be described and explained with additional specificity and detail through the use of the accompanying drawings.

FIG. 1A illustrates the basic concept of the present invention;

FIG. 1B illustrates further some of the concepts of the present invention;

FIG. 2 illustrates the basic hardware components according to an embodiment of the invention; and

FIG. 3 illustrates a method embodiment of the invention.

DETAILED DESCRIPTION OF THE INVENTION

Various embodiments of the invention are discussed in detail below. While specific implementations are discussed, it should be understood that this is done for illustration purposes only. A person skilled in the relevant art will recognize that other components and configurations may be used without parting from the spirit and scope of the invention. The present invention relates to resource reservations in the context of a cluster environment. The cluster may be operated by a hosting facility, hosting center, a virtual hosting center, data center, grid, cluster and/or utility-based computing environments.

Every reservation consists of three major components: a set of resources, a timeframe, and an access control list (ACL). Additionally, a reservation may also have a number of optional attributes controlling its behavior and interaction with other aspects of scheduling. A reservation's ACL specifies which jobs can use the reservation. Only jobs which meet one or more of a reservation's access criteria are allowed to use the reserved resources during the reservation timeframe. The reservation access criteria comprises, in one example, at least following: users, groups, accounts, classes, quality of service (QOS) and job duration. A job may be any venue or end of consumption of resource for any broad purpose, whether it be for a batch system, direct volume access or other service provisioning.

A workload manager, or scheduler, will govern access to the compute environment by receiving requests for reservations of resources and creating reservations for processing jobs. A workload manager functions by manipulating five primary, elementary objects. These are jobs, nodes, reservations, QOS structures, and policies. In addition to these, multiple minor elementary objects and composite objects are also utilized. These objects are also defined in a scheduling dictionary. A “job” as used herein is a general term in the sense that a job may include a number of subjobs which also may be referred to simply as jobs. Using the above example of a weather analysis job that runs on a cluster environment, consider that the weather analysis is a “job” but requires and will consume several different types of resources within the environment, such as data staging (disk space), processor time, bandwidth requirements and so forth. Each request for individual resources may be considered or termed a “job.” The present invention specifically relates to the scenario where there are at least two jobs that have an interrelationship in some fashion. For example, the data staging job needs to begin processing a certain amount of time prior to the start of the job that consumes processing time.

A workload manager may operate on a single computing device or multiple computing devices to manage the workload of a compute environment. The “system” embodiment of the invention may comprise a computing device that includes the necessary hardware and software components to enable a workload manager or a software module performing the steps of the invention. Such a computing device may include such known hardware elements as one or more central processors, random access memory (RAM), read-only memory (ROM), storage devices such as hard disks, communication means such as a modem or a card to enable networking with other computing devices, a bus that provides data transmission between various hardware components, a keyboard, a display, an operating system and so forth. There is no restriction that the particular system embodiment of the invention has any specific hardware components and any known or future developed hardware configurations are contemplated as within the scope of the invention when the computing device operates as is claimed.

We now turn to the details of the invention which relate to managing a compute environment in the context of maximizing usage of resources where two or more jobs require disparate but interrelated resources. FIG. 1A illustrates for 4 nodes within a cluster or a grid wherein a fixed time relationship between the jobs is created which allows specification and enforcement of a fixed time offset relationship 180 between each of the jobs resulting in at least one of a job overlap, juxtaposition or an offset-based relationship. An aspect of the invention involves establishing a forced relationship in space so that all jobs allocate the same resource objects while dedicating subsets of the resource's consumable resources to themselves. The resources mentioned herein relate to the various resources used within a cluster of computer nodes. For example, processing power, data storage, bandwidth in a network, licensing of software, RAM, and so forth. The example will relate to two resources of processor power and data storage but the invention may relate to any two different types of resource within the compute environment. Furthermore, the invention relates to at least two jobs but may involve two or more jobs requiring two or more disparate resources.

FIG. 1A shows a portion of a computing environment such as a cluster or a grid. The compute environment may have tens, hundreds, thousands (or more) processors or nodes within the environment. Illustrated therefore is a representative computing environment in which the concept of the invention may be applied. FIG. 1A shows four nodes 116, 118, 120, 122 each having two types of resources: computer processor resources (P) and disk storage/data staging resources (D). There may be any number of nodes in which the invention may be operable and the two types of resources are only illustrative. For example, other types of resources could also be shown in a similar fashion. Assume that a job 1 requires a certain number of hours of computer processing time. Also assume that there is an associated job 2 that requires some amount of disk space time. The relationship between these is flexible. In other words, there may be a master job that has two or more sub-jobs (jobs 1 and 2) or resource requests for processor time, disk space, etc. It is not a requirement in consuming resources on the compute environment that there be more than one job or resource request. However, it is the context of the present invention where disparate types of resources are interrelated. Jobs or resource requests that have a highest priority may be provided first access to all available resources.

FIG. 1A shows several features on node 1 116. This figure shows an analysis and some of the steps of the invention. Feature 126 may represent used or unavailable resources at a particular time. The horizontal axis in FIGS. 1A and 1B relates to time with, typically, the “0 hr” time as current time. Features 130, 132 and 102 represent available timeframes for job 1 to utilize the processor of node 1. Features 138, 140 and 136 represent available timeframes for utilizing disk space associated with job 2. These bars shown as associated with node 1 116 are not reservations but represent available time frames on which a reservation could be made. Thus, for job 1, available timeframes are shown at which the job could start include at 1 hr from the current time, two hours from the current time and three hours from the current time. Similarly, time frames in which the job 2 could start are shown in features 138, 140 and 136.

The graph of FIG. 1A is used to illustrate the resources (on the vertical axis) versus time (on the horizontal axis) for several nodes in the compute environment. It illustrates how the available timeframes for the jobs that utilize disparate types of resources will be processed to analyze the compute environment and maximize the resulting reservation of the compute environment to consume resources when the jobs process. Utilizing these features of the invention help to increase the efficiency of the compute environment.

Feature 150 represents a logical AND'ing of the available timeframes for processors and disk space and thus represents a listing of available timeframes for that node. Feature 152 represents a conversion of each availability timeframe into a start range. The system performs some analysis on the listing of available time ranges to determine given parameters such as when each job can start, and how long each job is, and other parameters, the actual available start ranges 152A, 152B for jobs may not be the same as the available timeframes 150. For example, to generate the start range, the system utilizes a wall time which is applied to the availability timeframes to generate the start range. The wall time relates to the amount of time that a particular resource will be required to process a job. Thus, if an availability time (130 FIG. 1B) runs an hour, but the wall time for that resource will be required is ˝ hour, then the start time associated with that particular availability timeframe must be for ˝ hour as is represented by 130A. Note the x's below several of the timeframes in 150 which illustrate that these are disqualified as time ranges for jobs to actually consume resources within the compute environment.

Node 2 118 is shown with unavailable resources 128 and available timeframes 154, 156, 158, 160 for a resource type of a processor for a job, and available timeframes for disk space at 162, 164 and 166 for another job. These timeframes may be for job 1 and job 2 respectively if these jobs will be utilizing more than one node. Feature 170 represents the available timeframes for node 2 118 which is an AND'ing of the available timeframes associated with the different resource types. Feature 172A, 172B and 172C represent the start range for node 2 which is the result of an analysis of when jobs can actually start and complete given all other relative parameters. Node 3 120 is shown with blocks 174 and 176 showing unavailable resources. As can be appreciated, other resource types may be added to this concept, such as bandwidth, RAM, and so forth. The use of the processors and disk space is illustrative only.

Having introduced some of the features of FIG. 1A, we further discuss the overall concept of the invention. The process involves selecting a resource (which may be a node as in node 1 116 of FIG. 1A, or a particular resource such as a processor or disk space etc.) and then traversing all of the fixed time relationship jobs associated with that resource, starting with a first job, and then processing a second job in comparison to the first, and then a third job and so forth until all the jobs have been processed. Then the next resource and processes the jobs associated with that resource and then blends the processing of the various resources each having its associated jobs. Then a global analysis is done to blend the job-level and resource-level analysis in order to generate a listing of acceptable resource at which to start jobs and the time frame at which to start those jobs. These nested processing loops enable a user to receive the maximized listing of resources and available times to start. Thus, when the system walks through the analysis for a resource (like a node), the system identifies the bounds, time and quantity of the resource or resources on the node. Then the system walks through again and focuses on the particular range and identifies the details about how the system came up with the details of availability and start times, etc. in the first place. After the analysis, if the system creates a reservation to process the job, the system knows that the reservation will enable the compute environment to adequately process the at least two jobs with their offsets and disparate resource types.

A user interface may include the ability of a user to select from the list of acceptable resources and starting times or the system may automatically pick a particular time such as the first timeframe at which the jobs can process. The global analysis provides a global resource availability range which represents availability of all nodes in the compute environment, such as a cluster, which can co-allocate all the resources associated with the jobs.

In one aspect of the invention, the available timeframes shift. When this shift occurs, all of the available time frames 130, 132, 102 are equally shifted in time by the offset. In our example, all of these available timeframes are shifted back in time ˝ hour. It does not matter which resource type is shifted and in which direction. In other words, the disk space available timeframes 138, 140 and 136 may be shifted to the right by the offset of ˝ hour. The only requirement is that the system track which availability timeframes associated with which resource type are shifted and which are not.

Returning to FIG. 1A and FIG. 1B, the start range 152A, 152B is shifted by a certain amount which is preferably the specified job offset 180. This shifting is illustrated in FIG. 1B. The job offset represents a different in time between the start of a first job associated with a first resource type and the start of a second job consuming a different resource type but that is connected to the first job. For example, first job may be job 130 that starts at 1 hour and job 138 that starts ˝ hour before job 130. In this case, the job offset would be ˝ hour. This job offset is an offset between any two job types. Therefore, it is not just between a job associated with disk space and a job associated with consuming processors. It may be an offset between any two associate jobs or resource requests of two resource types. FIG. 1B shows that all of the range list timeframes shift together. This is the preferable way to perform this shift. Each may separately be shifted as well or be shifted in different directions based on various parameters, such as, for example, the particular offset relationship between each range and an associated timeframe from a different resource. The shifted data whether it be from the start range or the available time frames for a particular job are then copied into a node range.

Next, a node range is generated which reflects available resources associated within a node. The copying of the shifted start range (or shifted availability timeframes or other data that may be shifted) into the node range is performed to a first job. For all subsequent jobs, the system logically AND's the respective job's start range with the node range. The start range in this case reflects, in one aspect of the invention, the start range for the respective subsequent job that is being processed. This may reflect the start range generated for each subsequent job which is AND'd together with the one node range or it may relate to different start ranges and different node ranges. Thus, in one aspect of the invention, the node range begins with being the start range copied from the first job. Then, each of the subsequence jobs, the start range for that particular job is AND'd with the node range and the node range thus becomes more constricted. Once all the jobs have been processed, the node range represents the range of available time all resources are available within a given node.

The system continues to process jobs within a node (or a resource) and then move to a next node (resource) in a compute environment. In this regard there are several nested loops of processing a node or a particular resource, and jobs or resource requests therein, and then move on to the next node or resource. Once all of the jobs associated with a resource or a node, such as all jobs on node 1 116, and in one aspect of the invention after all nodes have been processed according to the steps above, the system logically OR's the node range with a global range. The global range reflects the timeframes and resource quantities available to the overall grouping of jobs where all jobs can simultaneously consume their requested resources at the specified offsets on the same set of allocated resources. Thus, the global range may be looking at timeframes and resources available on a cluster, grid or other grouping of compute resources. From this OR'ing of the node range with a global range, the system can generate a list of acceptable resources on which to start jobs and the timeframe at which to start. Once that list is generated, the workload manager or other management software generates a reservation for resources according to the list of acceptable resources and associated job offsets.

FIG. 1C illustrates in another way the process of an aspect of the invention. As with FIGS. 1A and 1B, this shows two resources (processor and disk space) of the many resources that may be associated with a node 116 in a compute environment. The first row shows the availability timeframes for two jobs for resources on node 1. The features shown correspond to those of FIG. 1A. The next row shows start ranges for each availability time frame. For example, start ranges 130A, 132A and 102A correspond respectively to the availability time frames 130, 132 and 102 and represent the start range for each of the time frames. As mentioned above, the start range represents, in one aspect, a possible time frame at which the job could start. Thus, the start range is more restrictive than the availability range. The start ranges for the disk resource are shown as 138A, 140A and 136A.

The next row illustrates the shifting of the start ranges in time by, for example, the offset 180. Thus, start ranges 120A, 132A and 102A are all shifted to the left in time by the offset 180. Any time period may be chosen, and the other resource start ranges 138A, 140A and 136A could also be shifted by a time, such as time 180, to the right to match up. Whichever group of start ranges is shifted may be shifted together as a unit as shown.

The next row illustrates a node range 182A, 182B, 182C. This represents the range of available time all resources contained within a given node are available. This may be generated in several different ways such as through logical AND or OR operations associated with the various start ranges. For example, shown in the Node Range row of FIG. 1C is the logical AND'ing of the shifted start ranges for the resource of processor and the start ranges for the disk storage. There may be other ways to generate the node range. As can be appreciated, the node range represents a more constraining listing of times than the start ranges.

The next row 190 of FIG. 1C represents a global range 184A, 184B, 184C and 184D shows show the timeframes and resource quantities available to the overall grouping of jobs. Row 192 shows a global list 186A, 186B and 186C which show a list of acceptable resources on which to start jobs and the timeframe at which to start. This may be generated in several different ways. Shown in FIG. 1C is a logical OR'ing of a node range and the global range. Finally, row 194 shows a reservation 188 of resources. As mentioned herein, a user may select this reservation or it may be automatically generated according to a set of parameters. The reservation 188 reflects the generation of a reservation associated with the global listing 186B of available resources. With the reservation 188, the job that needs to be submitted to consume resources will be established with its necessary offset between disparate resources and be processed in an efficient manner.

We now turn to a general description of a computing device according to an embodiment of the invention. With reference to FIG. 2, an exemplary system for implementing the invention includes a general purpose computing device 200, including a processing unit (CPU) 220, a system memory 230, and a system bus 210 that couples various system components including the system memory 230 to the processing unit 220. The system bus 210 may be any of several types of bus structures including a memory bus or memory controller, a peripheral bus, and a local bus using any of a variety of bus architectures. The system may also include other memory such as read only memory (ROM) 240 and random access memory (RAM) 250. A basic input/output (BIOS), containing the basic routine that helps to transfer information between elements within the computing device 200, such as during start-up, is typically stored in ROM 240. The computing device 200 further includes storage means such as a hard disk drive 260, a magnetic disk drive, an optical disk drive, tape drive or the like. The storage device 260 is connected to the system bus 210 by a drive interface. The drives and the associated computer-readable media provide nonvolatile storage of computer readable instructions, data structures, program modules and other data for the computing device 200. The basic components are known to those of skill in the art and appropriate variations are contemplated depending on the type of device, such as whether the device is a small, handheld computing device, a desktop computer, or a computer server.

Although the exemplary environment described herein employs the hard disk, it should be appreciated by those skilled in the art that other types of computer readable media which can store data that is accessible by a computer, such as magnetic cassettes, flash memory cards, digital video disks, Bernoulli® cartridges, random access memories (RAMs) read only memory (ROM), and the like, may also be used in the exemplary operating environment. The system above provides an example server or computing device that may be utilized and networked with a cluster, clusters or a grid to manage the resources according to the principles set forth herein. In this case, the system aspect of the invention becomes the computing environment having multiple processors and nodes as is the case in a cluster or a grid. It is also recognized that other hardware configurations may be developed in the future upon which the method may be operable.

FIG. 3 illustrates the method embodiment of the invention. The method relates to scheduling multiple jobs with a fixed relationship in both time and space. A method consisting of two or more jobs where each job has its own duration, resource requirements, allocated resources and actions. Other parameters may also be associated with each job. The method establishes a fixed time relationship between the jobs which allows specification and enforcement of a fixed time offset relationship between each of the jobs resulting in a job overlap, juxtaposition or an offset-based relationship. Further, the method comprises the forced relationship in space so that all jobs allocate the same resource objects while dedicating subsets of the resource's consumable resources to themselves.

As shown in FIG. 3, the method comprises determining whether a job has a highest priority, and if yes, providing first access to all currently available resources (302), traversing a list of available resources (304), for each resource, traversing a list of fixed time relationship jobs (306). In this case, each resource may refer to a particular resource such as processor or disk space or may refer to a node such as node 1, 2 or 3 in FIG. 1A. The method further comprises obtaining a range list of available timeframes for each job(308), converting each availability range to a start range (310), shifting the resulting start range (or other data) in time by the specified amount of time such as by the job offset (312), for a first job, copying the resulting start range into a node range (314), for all subsequent jobs, logically AND'ing the start range with the node range (316), once all jobs have been processed, logically OR'ing the node range with a global range (318), and once all nodes have been processed, the resulting global range reflects the timeframes and resource quantities available to the overall grouping of jobs where all jobs can simultaneously consume their requested resources at the specified offsets on the same set of allocated resources (320). Based on the information, the method comprises generating a list of acceptable resources on which to start and the timeframe at which to start (322) and resource reservations can then be created (324).

In one aspect of the invention, the nested loops in the flowchart of FIG. 3 are that steps 306-318 are processed on the resource or node level, and steps 308-316 are nested within the node and relate to processing each job within the resource or node. Then starting at step 320 the method turns to a global view to further analyze the overall grouping of jobs and other resources available from which the method involves generating the list of available resources on which to start jobs and the timeframes at which to start.

A user interface is also involved in which an administrator or a user who submits jobs or some other user can graphically view the analysis. Therefore, there are various steps involved in the interaction with a user, such as presenting the generated list of acceptable resources on which to start jobs and available timeframes at which to start. There may be options such as an opportunity to pay an additional amount of money for an earlier time frame that is available but perhaps is outside of the scope of a particular usage package that the user or group has. For example, if a particular group has paid for a service that guarantees processing time within 3 hours of a request, and an available time based on the analysis above includes a time frame in 1 hour, that the user may opt to pay an additional amount (or be given the opportunity for free) to go ahead and select that start time of 1 hour. As another example of the user interaction, the user may be presented with available time frames from starting in 1 hour, 3.5 hours and 10 hours. The user then may be able to select and submit a request (say 10 hours) and the system will then establish the reservation for the resources according to the request. The system may alternately have an automatic approach where the user has a profile in which the earliest time is always selected, or the cheapest time, or the time that insures the job to be processed before 9:00 AM the next day, and so forth.

Embodiments within the scope of the present invention may also include computer-readable media for carrying or having computer-executable instructions or data structures stored thereon. Such computer-readable media can be any available media that can be accessed by a general purpose or special purpose computer. By way of example, and not limitation, such computer-readable media can comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to carry or store desired program code means in the form of computer-executable instructions or data structures. When information is transferred or provided over a network or another communications connection (either hardwired, wireless, or combination thereof) to a computer, the computer properly views the connection as a computer-readable medium. Thus, any such connection is properly termed a computer-readable medium. Combinations of the above should also be included within the scope of the computer-readable media.

Computer-executable instructions include, for example, instructions and data which cause a general purpose computer, special purpose computer, or special purpose processing device to perform a certain function or group of functions. Computer-executable instructions also include program modules that are executed by computers in stand-alone or network environments. Generally, program modules include routines, programs, objects, components, and data structures, etc. that perform particular tasks or implement particular abstract data types. Computer-executable instructions, associated data structures, and program modules represent examples of the program code means for executing steps of the methods disclosed herein. The particular sequence of such executable instructions or associated data structures represents examples of corresponding acts for implementing the functions described in such steps.

Those of skill in the art will appreciate that other embodiments of the invention may be practiced in network computing environments with many types of computer system configurations, including personal computers, hand-held devices, multi-processor systems, microprocessor-based or programmable consumer electronics, network PCs, minicomputers, mainframe computers, and the like. Embodiments may also be practiced in distributed computing environments where tasks are performed by local and remote processing devices that are linked (either by hardwired links, wireless links, or by a combination thereof through a communications network. In a distributed computing environment, program modules may be located in both local and remote memory storage devices.

Although the above description may contain specific details, they should not be construed as limiting the claims in any way. Other configurations of the described embodiments of the invention are part of the scope of this invention. For example, each of the processing steps above may involve various logical operations, such as AND'ing or OR'ing timeframes or other data together, or performing some other operation to move to the next step. Accordingly, the appended claims and their legal equivalents should only define the invention, rather than any specific examples given.

Referenced by
Citing PatentFiling datePublication dateApplicantTitle
US7568199 *Nov 12, 2003Jul 28, 2009Sap Ag.System for matching resource request that freeing the reserved first resource and forwarding the request to second resource if predetermined time period expired
US7680970 *Oct 22, 2004Mar 16, 2010Fisher-Rosemount Systems, Inc.Method and system for batch process arbitration in a process control system
Classifications
U.S. Classification718/104
International ClassificationG06F9/46
Cooperative ClassificationG06F9/5027, G06F9/46, G06F9/5038, G06F9/4881
European ClassificationG06F9/50A6E, G06F9/50A6, G06F9/48C4S
Legal Events
DateCodeEventDescription
Jun 15, 2010ASAssignment
Free format text: CHANGE OF NAME;ASSIGNOR:CLUSTER RESOURCES, INC.;REEL/FRAME:24539/242
Effective date: 20090827
Owner name: ADAPTIVE COMPUTING ENTERPRISES, INC.,UTAH
Free format text: CHANGE OF NAME;ASSIGNOR:CLUSTER RESOURCES, INC.;REEL/FRAME:024539/0242
Owner name: ADAPTIVE COMPUTING ENTERPRISES, INC., UTAH
Feb 9, 2006ASAssignment
Owner name: CLUSTER RESOURCES, INC., UTAH
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:JACKSON, DAVID B.;REEL/FRAME:017148/0160
Effective date: 20050310